Interactive: Unravelling air pollution in Asia

Country/area: United Kingdom

Organisation: The Third Pole

Organisation size: Small

Publication date: 28/07/2021

Credit: Disha Shetty, Julia Janicki, Marta Portocarrero, Lou Del Bello


Disha Shetty is a science journalist currently based in Pune, India. She writes mainly about public health, environment, gender and the intersections of these issues.

Julia Janicki is a freelance data journalist and data viz developer focusing on environmental issues, biodiversity conservation and humanitarian issues. She’s from Taiwan and the US and is currently based in Paris.

Marta Portocarrero is a Graphics Editor with China Dialogue Trust, based in London. She specialises in video and data journalism.

Lou Del Bello is a Special Projects Editor with The Third Pole, specialising in climate change and energy in South Asia.

Project description:

The project looks at the sources and impact of air pollution across South Asia, with a particular focus on public health, a sensitive transboundary issue in an area of frequent border tensions. Monthly PM 2.5 data – from US Air Now platform and IQAir – is used to visualise the spread and seasonal fluctuation of air pollution in major cities in the Hindu Kush Himalayan region from 2018 to 2020. The data analysis acts as a springboard to investigate the health impacts on the region’s population and its causes. It includes video testimonies from people in the highest polluted cities.

Impact reached:

Our work with public but previously unexplored datasets helps break the longstanding blame game that year after year sees countries pointing at each other when the air gets toxic. Instead, we prove that regional air quality trends are often driven by local sources of pollution. The project also helped raise awareness of the fact that much of the air pollution in the region was from local sources.It also sparked debate (mainly on social media) and pointed out possible future solutions, and generally stimulated an open debate on the issue. The project was published a few months ahead of the COP26 summit in Glasgow where air pollution and its public health impacts were discussed. A conversation ahead of the pivotal event was the perfect segue to amplying local issues ahead of the global summit, giving space to the concerns of the developing world breathing in the heavily polluted air.  

Techniques/technologies used:

We collected open source PM2.5 hourly data of 21 cities over the past three years. We used Microsoft Excel to clean the datasets and Dynamic Tables to calculate monthly averages from the hourly data. We then used HTML, CSS and JavaScript to build an interactive page on the web. The base map was styled using Mapbox studio, and Mapbox GL JS (JavaScript library) was used to render the Mapbox onto the webpage, as well as to add points to the map and add filtering functionalities. The overall line charts, as well as the city-specific line chart showing air quality over time were built with D3, another JavaScript library.

On the writing side we made a conscious attempt to balance male and female voices as well as to include voices from multiple countries. Our sources were based across India, Pakistan, Nepal and the United States and this was a result of a deliberate attempt to highlight local voices as well as global research.

What was the hardest part of this project?

The most difficult part was finding accurate data sources for air pollution in the region. While the datasets are made available by some countries’ governments, many data points were missing. The solution came from the US Air Now platform, which places air quality monitors at US embassies in different cities. Another challenge was to clean it and export the data in a format that could be easily visualised. For this, we used Microsoft Excel dynamic tables, which helped extract monthly averages for each city from 2018 to 2020. In terms of development and design, a main challenge was to find an alternative way to display the countries without showing the borders and avoid controversy over many disputed borders across the region. Originally, the average air quality index was displayed as a choropleth map in the background, with each country being coloured based on the index. The best alternative way we found to display these values was to display them as circles, centred on the latitude / longitude centroid of each country.

While writing the story we reached out to a team who had done a global study on the sources of air pollution and asked them for insights specific to the region given the absence of the region-specific study. The absence of data or the below par quality of it when present is always a major challenge of reporting on any developing country. This region also doesn’t maintain good public health records on deaths and the causes of death. This means attributing any mortality to air pollution is tricky and while we did mention that in the story, communicating this complex science in a way that is both accessible and nuanced was tricky.

What can others learn from this project?

Other journalists can use this project as an example of how to visualise air pollution, which can be a very abstract concept. The project can also inspire them to work on data journalism projects and encourage them to be inquisitive when data trends show realities that differ from commonly held perceptions.

The intersection of air pollution and public health is very visible and underreported. This project can also act as a guide to how this intersection can be covered in an engaging way. And finally, given the story is rooted in a developing region, it raised the key issue of the lack of communication between countries. This story can act as an example to other journalists on an example of effective cross border storytelling that focuses on a common problem. 

Project links: